ABSTRACT Activation induced deaminase (AID) is essential for both immunoglobulin (Ig) somatic hypermutation (SHM) and class switch recombination (CSR) in mature B cells. AID deaminates dC to dU. Using the base excision repair (BER) pathway, many (but not all) AID induced dU bases can be excised by a uracil DNA glycosylase (UNG) leaving an abasic site that can then be replicated over by error prone polymerase to produce both transition and transversion mutations. Genetic studies show that UNG deficiency in mice and humans leads to loss of class switch recombination (CSR) and impaired somatic hypermutation (SHM). U:G mismatches are also substrates for mismatch repair (MMR) MSH2/MSH6 binding which in turn recruit the MutL? homologues, PMS2-MLH1. PMS2 nicks the DNA and contributes to the induction of double strand breaks in switch regions, the AID targets during CSR. Although MSH2/MSH6 are critical for the induction of A/T mutations, PMS2-MLH1 are not active in SHM. This raises the intriguing question of which nuclease contributes the critical nick to initiate A/T mutagenesis during SHM. There are four uracil DNA glycosylases, UNG, SMUG1, TDG, and methyl binding domain 4 (MBD4), that are capable of recognizing and removing dU in U:G mismatches. Evidence indicates that TDG cannot substitute for UNG and SMUG1 plays little natural role in these processes since it is poorly expressed in activated B cells. MBD4 protein was originally discovered by virtue of its interaction with MLH1. We were intrigued by the functional association of MBD4 and AID in the context of active DNA demethylation in zebrafish. However, previous studies indicated no phenotype with regard to CSR and SHM in Mbd4 deficient mice. Based on our observation of alternative Mbd4 splice variants with open reading frames which could enable expression of the C-terminal end of MBD4 even when 5'-Mbd4 is deleted, we constructed CH12 cells and mice deleted 3'- Mbd4. Strikingly we found in new studies that CSR was significantly impaired in 3'-Mbd4 deficient CH12 cells. Furthermore, SHM was essentially abolished in 3'-Mbd4 deficient mice. We are currently characterizing 3'-Mbd4 deficient mice with respect to CSR and SHM. We propose to construct a new mouse that expresses a catalytically dead MBD4 to definitively determine the role of this uracil glycosylase in SHM, CSR and cancer.